important that we find a large number of lower animal forms to be
still represented in the course of man's embryonic development. In
these cases we may draw our conclusions with the utmost security as to
the nature of the ancestral form from the features of the form which
the embryo momentarily assumes.

To give a few examples, we can infer from the fact that the human ovum
is a simple cell that the first ancestor of our species was a tiny
unicellular being, something like the amoeba. In the same way, we
know, from the fact that the human foetus consists, at the first, of
two simple cell-layers (the gastrula), that the gastraea, a form with
two such layers, was certainly in the line of our ancestry. A later
human embryonic form (the chordula) points just as clearly to a
worm-like ancestor (the prochordonia), the nearest living relation of
which is found among the actual ascidiae. To this succeeds a most
important embryonic stage (acrania), in which our headless foetus
presents, in the main, the structure of the lancelet. But we can only
indirectly and approximately, with the aid of comparative anatomy and
ontogeny, conjecture what lower forms enter into the chain of our
ancestry between the gastraea and the chordula, and between this and
the lancelet. In the course of the historical development many
intermediate structures have gradually fallen out, which must
certainly have been represented in our ancestry. But, in spite of
these many, and sometimes very appreciable, gaps, there is no
contradiction between the two successions. In fact, it is the chief
purpose of this work to prove the real harmony and the original
parallelism of the two. I hope to show, on a substantial basis of
facts, that we can draw most important conclusions as to our
genealogical tree from the actual and easily-demonstrable series of
embryonic changes. We shall then be in a position to form a general